Senior AgentOps Architect - LLM/MLOps & AI Observability

TechDigital Group; Santa Clara (CA); On-site; USD 180,000 - 240,000; Full time

Job summary

TechDigital Group is seeking a Senior Subject-Matter Expert for AgentOps/LLMOps to own the operations, observability, and lifecycle management of AI agents in production. You will ensure reliability, safety, and cost-efficiency while driving KPI performance across the Agent Factory.

You will define the AgenticOps framework, establish continuous evaluation, and implement strong observability. Collaboration with DevOps and AI/Data SMEs is essential to close the build-deploy-operate-improve loop.

Qualifications

• Extensive LLMOps/AgentOps experience in production environments.
• Hands-on with Google Cloud runtimes and observability tooling.
• Strong knowledge of model governance, safety, and HITL testing.
• Proficient Python with deep understanding of agent lifecycle and governance.

Responsibilities

• Define and operate AgenticOps framework including registry and rollout controls.
• Establish continuous evaluation and monitoring of agent safety and cost.
• Implement observability/tracing for multi-agent systems and post-release processes.
• Design HITL supervision and automated pre-production simulations.
• Track KPIs via dashboards and governance mechanisms; optimize runtimes and budgets.
• Collaborate with DevOps and AI/Data SMEs to improve end-to-end lifecycle.

Skills

LLMOps

AgentOps

MLOps

Python

GCP

Observability

SRE

Cost governance

Education

Google Cloud Professional (ML/DevOps) certification

Tools

Vertex AI

Agent Engine

BigQuery

Looker

Job description

TechDigital Group is seeking a Senior Subject-Matter Expert for AgentOps/LLMOps to own the operations, observability, and lifecycle management of AI agents in production. You will ensure reliability, safety, and cost-efficiency while driving KPI performance across the Agent Factory.

You will define the AgenticOps framework, establish continuous evaluation, and implement strong observability. Collaboration with DevOps and AI/Data SMEs is essential to close the build-deploy-operate-improve loop.